MMMU-Benchmark/MMMU

• highabout#1

  Reposition the 'About' description to emphasize the dataset

  原因：

  当前

  This repo contains evaluation code for the paper "MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI"

  复制粘贴的修复

  复制

  MMMU is a massive, multi-discipline multimodal understanding and reasoning benchmark dataset, featuring 11.5K college-level questions for evaluating expert AGI across STEM and humanities fields. This repository provides the evaluation code.

  待办完成忽略
• mediumtopics#2

  Add specific dataset-related topics

  原因：

  当前

  computer-vision, deep-learning, deep-neural-networks, evaluation, foundation-models, large-language-models, large-multimodal-models, llm, llms, machine-learning, multimodal, multimodal-deep-learning, multimodal-learning, multimodality, natural-language-processing, question-answering, stem, visual-question-answering

  复制粘贴的修复

  复制

  computer-vision, deep-learning, deep-neural-networks, evaluation, foundation-models, large-language-models, large-multimodal-models, llm, llms, machine-learning, multimodal, multimodal-deep-learning, multimodal-learning, multimodality, natural-language-processing, question-answering, stem, visual-question-answering, benchmark-dataset, multimodal-dataset, ai-benchmark-dataset

  待办完成忽略
• lowreadme#3

  Refine README's initial paragraph to include the dataset

  原因：

  当前

  This repo contains the evaluation code for the paper "MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark" and "MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI"

  复制粘贴的修复

  复制

  This repository provides the evaluation code and the massive multimodal dataset for the papers "MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark" and "MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI"

  待办完成忽略

本次扫描解析到的品类 GEO 通道：google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• 品类问题

  How to benchmark large multimodal models for advanced reasoning and multi-discipline understanding?

  你：第 1 位
  AI 推荐顺序：

  1. MMMU ← 你
  2. MME
  3. POPE
  4. MathVista
  5. ScienceQA
  6. VQAv2
  7. GQA
  查看 AI 完整回答
• 品类问题

  What are robust evaluation datasets for college-level multimodal AI performance across STEM fields?

  你：未被推荐
  查看 AI 完整回答

• Metadata completeness

  pass

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of MMMU-Benchmark/MMMU?

  pass
  ✓AI 明确点名了 MMMU-Benchmark/MMMU

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• If a team adopts MMMU-Benchmark/MMMU in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI 明确点名了 MMMU-Benchmark/MMMU

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？

• In one sentence, what problem does the repo MMMU-Benchmark/MMMU solve, and who is the primary audience?

  pass
  ✓AI 明确点名了 MMMU-Benchmark/MMMU

  AI 的回答可能信誓旦旦却是错的。请按事实核对：技术栈、目标人群、差异化点是不是和你实际的对得上？